OpenAI

OpenAI provides both Whisper-based transcription and GPT-based summarization, so a single API key covers both roles in Quendian.

Role	Models
Transcription	`whisper-1`, `gpt-4o-transcribe`, or `gpt-4o-mini-transcribe`
Summarization	`gpt-4o-mini`

Steps

1. Create an API key

Go to platform.openai.com/api-keys and click Create new secret key.

Key format: sk-proj-… (new project keys) or sk-… (legacy keys)

Copy the key immediately — OpenAI shows it only once.

2. Open Preferences in Quendian

Navigate to Quendian → Preferences. The same OpenAI API key works for both Transcription and Summarization — paste it into the role you want to configure first; you can paste it into the other role’s pane afterward.

3. Pick a transcription model

OpenAI offers three transcription presets. Pick one based on what matters most for your meetings.

Whisper-1

The original Whisper model. Word-level click-to-jump is available only with this preset — click any word in the transcript to jump audio playback to that moment. Strong on clean audio.

GPT-4o Transcribe

OpenAI’s newer transcription model. Generally cleaner output than Whisper-1, especially on accented English and imperfect mics. Segment-level navigation only — no word-level click-to-jump.

GPT-4o Mini Transcribe

A lighter variant of GPT-4o Transcribe. Same segment-level output, smaller model behind the scenes.

4. Configure summarization

Quendian Preferences — Summarization pane with the OpenAI preset selected

Navigate to Preferences → Summarization, pick the OpenAI preset, and paste the same key. Quendian uses gpt-4o-mini for summaries by default.

Same key, both roles

A single OpenAI API key authenticates both Transcription and Summarization. Paste it into both Transcription → OpenAI and Summarization → OpenAI. Quendian makes separate API calls per role — no extra setup required.

Monitor usage and pricing

For current rates, see OpenAI’s pricing page. Track your usage at platform.openai.com/usage.